intro-dst

Avoiding “Deaths by PowerPoint”
Data Storytelling and Python to the rescue
Sebastián Flores, PyCon US 2024, May 20th

Who

Sebastián Flores aka sebastiandres
Chief Data Officer at uPlanner
Member of Python Chile
Collaborator to Streamlit and Vizzu
http://sebastiandres.xyz

Agenda

Never
kill
the
suspense

Rule #1 of Storytelling: Always raise the stakes

Storytelling?

It just works

Data Storytelling:

Using storytelling tricks to create presentations that will be remembered and that create impact

We are only as good as our communication skills. If you’re really really good, the god of anything, but cannot articulate your ideas and communicate effectively, you’ll only be as good as some bad with great communication skills. Don’t make the mistake. Learn to communicate effectively. [Visual of a lake? A multiplication? A lever!] Public speaking is a lever. Think of Sherlock Holmes. We can admire his prodigious mind and memory only because of the work of Doctor John Watson. It’s not enough to crack the case and solve the mystery. We need to be able to collect the important facts and to tell an engaging story. Same goes for data science. It’s not enough to know the latest library or model the problem properly, it’s all about distilling the important fact that your boss needs to know to make the important decisions.

DSt is the new DSc

The best example

Can a company convince millions of people to share statistics on social media?

Skills vs Tools

Skills for Data Storytelling

Data Storytelling?

It’s easy to consume but hard to prepare

About stories

Stories have a beginning, a climax and a conlusion.

One of the secrets of Storytelling is that all stories have “The hero´s journey” - something studied by Joseph? Campbel on the ??, which sort of says that there is a common structure: normal world - conflict and call to adventure - … - resolution - return the normal world. Why should you care about this? This structure, this pattern, works. It’s engaging and the brain cannot avoid caring about the end. It creates an open loop, and people want to closure. It’s a pattern that works. If you have to remember only 1 thing from this talk, remember this: The audience for your presentation, treat each one of them as a hero on their own. They have their own inner world, and your presentation disrupts that world. You share with them a knowledge, something that SHOULD change how they see the world, the tools they have, anything. At least one phrase to be remembered. If nothing changes for the audience, they wont remember anything a couple of days or even hours after your talk. Memory is very selective. That’s why brains evolved to like stories. Now, how to connect this to python and tools?

Brains are so hackable!

Peak-end rule:

We remember the most intense moment and the final emotion.

There is a known psycology effect. You remember something: How it ended What was the most intense emotion This is key. Always build your presentation around the big ending you want. Work backwards. Create the five ending slides. If anything goes bad, you could still present only those 5 slides. Make sure that’s the ending you want and your audience deserves. Only then work backwards to the rest of the presentation. The peak–end rule is a psychological heuristic in which people judge an experience largely based on how they felt at its peak (i.e., its most intense point) and at its end, rather than based on the total sum or average of every moment of the experience. This is why we don’t remember the queue of the rollercoaster, and do remmeber the thrill! https://sketchplanations.com/the-peak-end-rule Image from

The hard part of Data Storytelling

How to manipulate without being manipulative?